Help needed with fsolve problem

[Terry Duell]

Hello All,
I have run into some difficulties in solving a set of simultaneous  
equations, using fsolve.
I have known (measured) accelerations from an array of accelerometers, and  
know the geometry.
What I want to do is derive the variables, shown below in the example code  
snippet.
I started with the approach shown in the example, coding the projections  
of each r, which produces results. (note this example code may not have  
correct maths)
=====start code======
% the variables
%       x(1) = theta, pitch angle
%       x(2) = theta-dot, pitch velocity
%       x(3) = theta-dotdot, pitch acceleration
%       x(4) = phi, roll angle
%       x(5) = phi-dot, roll velocity
%       x(6) = phi-dotdot, roll acceleration
%
% the solution based on setting local origin at accelerometer 0
% set up the function
        fcn =@(x) [a0z-r1z*cos(x(1))*x(3)-r1z*sin(x(1))*x(2)*x(2)-a1z,
                   a0z+r2z*cos(x(1))*x(3)+r2z*sin(x(1))*x(2)*x(2)-a2z,
                   a0z+r3z*cos(x(4))*x(6)+r3z*sin(x(4))*x(5)*x(5)-a3z,
                   a0z+r4z*cos(x(4))*x(6)+r4z*sin(x(4))*x(5)*x(5)-a4z,
                   ...etc...];


        [x info]= fsolve(fcn, [0 0 0 0 0 0]);
=======end code=======

What I would like to do is apply a rotation matrix to derive the projected  
distances, as per the following example, and use those projected distances  
instead of manually coding each

=======start code======
rotx = [1,0,0;0,cos(theta),-sin(theta);0,sin(theta),cos(theta)];
roty = [cos(theta),0,sin(theta);0,1,0;-sin(theta),0,cos(theta)];
rotz = [cos(theta),-sin(theta),0;sin(theta),cos(theta),0;0,0,1];
r = [rx,ry,rz];
rrotx = rotx*r';
newrz = rrotx(3,:);
=======end code========

The problem I have is that I can't see how to incorporate the rotation  
matrix steps into the set of simultaneous equations to pass to fsolve,  
such that x(1) and x(4) end up in the solution.

I should add, just in case it isn't clear, all the x variables and the  
a1z, a2z etc, are 1D arrays.

Any advice on this will be much appreciated.

Cheers,
--
Regards,
Terry Duell